Objective: To investigate the factors associated with subtherapeutic serum magnesium levels in order to prevent eclampsia in term pregnant women with severe pre-eclampsia. Methods : This case-control study included 200 term pregnant women with severe pre-eclampsia who received magnesium sulfate for eclampsia prophylaxis. These patients were randomly allocated into case and control groups. Experimental cases included 100 women whose serum magnesium level did not reach the therapeutic level (<4.8 mg/dL) at 2 hours after initial administration, whereas the controls were 100 women whose serum magnesium levels reached therapeutic levels (4.8-8.4 mg/dL). Data from the medical records, including baseline characteristics, sign and symptoms, laboratory findings and pregnancy outcomes, were extracted for univariate and multivariate analyses. Results: Only two factors, pre-pregnancy body mass index (BMI) and serum creatinine level, showed significant differences between the two groups. Pre-pregnancy BMI greater than 25 kg/m 2 increased the risk of subtherapeutic serum magnesium level by 56% compared with normal pre-pregnancy BMI (adjusted OR 1.56, p=0.019), whereas pre-pregnancy BMI less than 18 kg/m 2 decreased the risk by 80% (adjusted OR 0.2, p=0.02). Also, serum creatinine levels greater than 0.9 mg/dl decreased the risk by 98.7% (adjusted OR 0.013, p<0.001). No significant difference in pregnancy outcomes was noted in either group. Conclusion: Pre-pregnancy BMI greater than 25 kg/m 2 increased the risk, whereas pre-pregnancy BMI less than 18 kg/m 2 decreased the risk of subtherapeutic serum magnesium levels. A serum creatinine level greater than 0.9 mg/dl was another factor that decreased the risk.